AJCN North Carolina Research Campus
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Schienkiewitz, A.
Right arrow Articles by Boeing, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Schienkiewitz, A.
Right arrow Articles by Boeing, H.
Agricola
Right arrow Articles by Schienkiewitz, A.
Right arrow Articles by Boeing, H.
American Journal of Clinical Nutrition, Vol. 84, No. 2, 427-433, August 2006
© 2006 American Society for Nutrition

ORIGINAL RESEARCH COMMUNICATION

Body mass index history and risk of type 2 diabetes: results from the European Prospective Investigation into Cancer and Nutrition (EPIC)–Potsdam Study1,2,3

Anja Schienkiewitz, Matthias B Schulze, Kurt Hoffmann, Anja Kroke and Heiner Boeing

1 From the Department of Epidemiology, German Institute of Human Nutrition Potsdam–Rehbrücke, Germany (AS, MBS, KH, and HB), and the Department of Nutrition and Health, Research Institute of Child Nutrition, Dortmund, Germany (AK)

2 The follow-up of the EPIC–Potsdam Study was supported by a grant from the Deutsche Krebshilfe (70-2488-HAI) and by the European Union (SOC 95 201408 OSF02). Recruitment was supported by a grant from the Federal Ministry of Research and Technology (01 EA 9401).

3 Reprints not available. Address correspondence to A Schienkiewitz, German Institute of Human Nutrition, Department of Epidemiology, Arthur-Scheunert-Allee 114-116, 14458 Nuthetal, Germany. E-mail: anja.schienkiewitz{at}mail.dife.de.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Background: Obesity and increases in body weight in adults are considered to be among the most important risk factors for type 2 diabetes.

Objective: The objective was to evaluate and compare the associations between weight changes during 2 different periods of adult life and the risk of type 2 diabetes and age at diagnosis.

Design: The study included 7720 men and 10 371 women from the European Prospective Investigation into Cancer and Nutrition (EPIC)–Potsdam Study with information on weight history; 390 men and 303 women of these participants received a clinical diagnosis of type 2 diabetes during 7 y of follow-up. Multivariate Cox regression models were used to estimate the relative risk (RR) of weight changes between ages 25 and 40 y and ages 40 and 55 y.

Results: RR estimates in men and women were slightly higher for each unit of BMI gain between ages 25 and 40 y [men: 1.25 (95% CI: 1.21, 1.30); women: 1.24 (1.20, 1.27)] than between ages 40 and 55 y [men: 1.13 (1.10, 1.16); women: 1.11 (1.08, 1.14)]. Severe weight gain between ages 25 and 40 y was associated with a higher diabetes risk in men (1.5 times) and in women (4.3 times) than were stable weight in early adulthood and weight gain in later life, and it resulted in an average lower age at diabetes diagnosis in men (5 y) and in women (3 y).

Conclusion: Weight gain in early adulthood is related to a higher risk and earlier onset of type 2 diabetes than is weight gain between 40 and 55 y of age.

Key Words: Weight change • body mass index • obesity • age at diabetes diagnosis • onset of type 2 diabetes


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Results from metabolic and epidemiologic studies provide strong evidence that obesity is causally related to type 2 diabetes (1–11). Many studies have reported associations between body mass index (BMI; in kg/m2) and type 2 diabetes in men (2, 7–9, 11–19) and women (2, 9, 10, 14–16, 18, 20). Investigations focusing on weight change and type 2 diabetes showed that, besides obesity per se, an increase in body weight of 3–20 kg is associated with an elevated risk of incident type 2 diabetes (2, 7, 9–12, 15, 18, 19, 21, 22). Early obesity and almost any weight gain after adolescence are risk factors for type 2 diabetes (7, 10). Moreover, the duration of obesity seems to be a significant risk factor for type 2 diabetes, independently of current degree of obesity (6, 19, 23).

An inverse linear relation was found between BMI and age at diabetes onset (24). Adults with early diagnosed diabetes were more obese and more likely to be female than were adults with a later onset of type 2 diabetes (25).

Although many studies have reported associations between weight gain and risk of type 2 diabetes, no study explicitly quantified whether body weight changes during different periods in adult life are differently related to risk of type 2 diabetes and how much such weight changes influence the age at diabetes diagnosis. This information is urgently needed because a large segment of the population is starting to gain weight early in adult life, after settling into an occupation or family life, which is contributing to the obesity epidemic. In Germany, the largest increase in the prevalence of obesity in men is seen between ages 20 and 40 y and in women between ages 30 and 40 y (26). The present investigation examined the effect of body weight gain in early life on the age at diabetes diagnosis and compared 2 different periods in adult life with regard to the association between weight change and diabetes risk.


   SUBJECTS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Study subjects
The European Prospective Investigation into Cancer and Nutrition (EPIC)–Potsdam Study is part of the multicenter prospective cohort study EPIC, which was designed to investigate the association between nutrition and cancer and other chronic diseases (27). In Potsdam, Germany, 27 548 subjects (n = 16 644 women aged 35–65 y and 10 904 men aged 40–65 y) from the general population were recruited from 1994 until 1998. The baseline examination included anthropometric measurements, blood samples, a self-administered validated food-frequency questionnaire, and a personal interview and questionnaires on prevalent diseases and sociodemographic and lifestyle characteristics. For the collection of follow-up information on current medication use and newly developed chronic diseases, including type 2 diabetes, questionnaires are mailed to the study participants every 2–3 y. Response rates for completely filled in follow-up questionnaires at follow-up rounds 1, 2, and 3 were 96%, 95%, and 91%, respectively. For this study, we also considered questionnaires sent out within the 4th follow-up round until 31 January 2005, of which 90% were returned by 31 August 2005. The recruitment, follow-up procedures and measures of quality control were previously described in detail (28–30).

Subjects with prevalent diabetes at baseline (n = 1443)—based on self-reported diagnosis, medication use, or dietary treatment—and subjects with self-reported diabetes during follow-up but without physician confirmation were excluded (n = 503). Furthermore, we excluded subjects with no follow-up information (n = 656) and missing confounder information at baseline for the following characteristics: BMI (n = 212), smoking (n = 17), and alcohol use (n = 17); we also excluded subjects aged ≤40 y at baseline (n = 4863), subjects with missing weight at ages 25 y (n = 1751) and 40 y (n = 6127), and subjects with implausible BMI changes, ie, a gain of >25 or a loss of > 15 during each weight-change period) (n = 337). After these exclusions, 7720 men and 10 371 women remained for analyses.

Exposure variables
Anthropometric measurements such as body weight and height were measured during the baseline examination by trained staff members following standardized procedures while the study participants were wearing no shoes and only underwear (31). During the personal computer-based interview, the participants were retrospectively asked for their weight at ages 25 and 40 y (28). A high degree of reproducibility of self-reported past body weight at ages 25 and 40 y between 2 interviews administered 1 y apart was observed in a substudy of the EPIC-Potsdam Study. Reproducibility of weight recall was within ±3 kg for 75.8% at age 25 y and 81.7% at age 40 y (32).

BMI calculated at ages 25 and 40 y and at baseline as reported or measured weight (kg) was divided by height squared (in m); the latter was measured only at baseline.

Lifestyle characteristics
Information on educational attainment, smoking, and alcohol consumption history were assessed with a self-administered questionnaire and a personal interview. History of smoking intensity was calculated as the average number of cigarettes, cigars, and pipes smoked per day at ages 20, 30, 40, and 50 y and at baseline. History of alcohol consumption was considered to be alcohol intake (g/d) from alcoholic beverages at ages 20, 30, and 40 y and at baseline (33). Sports activities were calculated as the average time spent per week during the 12 mo before baseline recruitment.

Identification of type 2 diabetes
Potentially incident cases of diabetes were identified on the basis of self reports of a diabetes diagnosis, use of diabetes-relevant medication (eg, insulin, sulfonylurea), or dietary treatment for diabetes. All potentially incident cases were verified by questionnaires that were mailed to the treating physicians, who were asked for the date and type of diagnosis, the diagnostic tests used, and the methods of treatment. Only cases with a physician diagnosis of type 2 diabetes [International Classification of Disease, 10th edition (ICD10): E11] and a diagnosis date after the baseline examination were considered as confirmed incident cases.

Statistical analysis
We used Cox proportional hazards models for type 2 diabetes to estimate adjusted hazard ratios and 95% CIs for changes in BMI. Age was used as the primary time variable in all models, with entry time defined as the subject's age at recruitment and exit time as the date of diagnosis of diabetes (ICD10: E10, E11, E13, and E14), death, or return of the last follow-up questionnaire. BMI at age 25 y and weight changes between ages 25 and 40 y and between 40 and 55 y were modeled simultaneously as 3 independent risk factors. We calculated BMI changes between ages 25 and 40 y and between age 40 y and the age at the time of the baseline examination. Because the duration between age 40 y and the time of the baseline examination varied across study participants and was on average shorter (median = 13 y) than the 15-y period between ages 25 and 40 y, we estimated the BMI change for a standardized 15–y period after age 40 y by linear regression. BMI at baseline can be decomposed and formally written as the sum of 3 components: BMI at age 25 y, BMI change between ages 25 and 40 y, and BMI change between age 40 y and age at the time of the baseline examination. This alternative parameterization is often applied in life course epidemiology (34). The total BMI change between age 25 y and age at the time of the baseline examination was calculated by subtracting the BMI at age 25 y from the BMI at age 55 y. Because, for the younger participants, the short period of time between age 40 y and the time of the baseline examination may have resulted in unreliable estimates of weight changes between ages 40 and 55 y, we restricted an additional analysis to those 4852 men and 6538 women who were older than 50 y at the time of the baseline examination (n = 552 cases).

Mean percentage BMI change was calculated as the ratio of the change within a period to the BMI at the beginning of that period multiplied by 100. We further grouped men and women into 3 weight-change categories: 1) loss or stable (loss or gain of <1 BMI unit over 15 y), 2) moderate gain (gain of 1.0–4.0 BMI units), and 3) severe gain (gain of >4.0 BMI units). For a person 1.70 m in height, a 1-unit increase in BMI corresponds to a weight gain of 2.9 kg.

We used information on covariates obtained from the baseline questionnaires and the interview in a multivariate analyses, including history of smoking status (never, exsmoker, or current smoker), history of alcohol consumption (0, 0.1–5.0, 5.1–10.0, 10.1–20.0, 20.1–40.0, or >40.0 g/d), physical activity (0, ≤4, or >4 h/wk), and educational level (less than high school, high school, more than high school). Hazard ratios for BMI change within different periods were tested to be significantly different from zero with Wald's test.

We evaluated potential effect modifications modeling interaction terms for baseline characteristics and sex with respect to overweight by using the BMI cutoff of 25 proposed by the World Health Organization and other institutions (35, 36). Tests for linear trend were performed by using anthropometric characteristics, physical activity, and alcohol consumption as the independent variables (continuous variables) in a linear regression model, and a test of independency (Cochran-Armitage trend test) was performed for categorical variables.

All statistical analyses were performed with SAS release 9.1 (SAS Institute, Cary, NC). All statistical tests were two-sided, and P values <0.05 were considered statistically significant.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
The baseline characteristics of the 7720 men and 10 371 women involved in our analyses are summarized in Table 1Go. BMI at ages 25, 40, and 55 y differed for both sexes and all categories of BMI at age 25 y. The mean percentage of BMI change between ages 25 and 40 y was highest (10.2%) for men with a BMI at age 25 y that was <23, 7.5% for men with a BMI between 23.5 and 25.0, and 5.5% for overweight men at age 25 y; the corresponding values for women were 9.1%, 6.7%, and 5.5%, respectively. Women had a higher percentage of BMI change from ages 40 to 55 y in all categories of BMI at age 25 y (14.1%, 11.8%, and 11.7%, respectively), whereas the percentage of BMI change for men increased slightly by 9.7%, 8.0%, and 7.5% compared with the earlier period. Men and women with a BMI < 23 at age 25 y had the highest percentage of BMI change between ages 25–40 y (men: 10.2%; women: 9.1%) and ages 40–55 y (men: 9.7%; women: 14.1%) compared with subjects with a BMI ≥ 23.0, although the absolute BMI at ages 40 and 55 y was lowest for men and women with a BMI < 23 at age 25 y.


View this table:
[in this window]
[in a new window]
  		TABLE 1 History of anthropometric measures and baseline characteristics by sex and BMI at age 25 y: European Prospective Investigation into Cancer and Nutrition (EPIC)–Potsdam Study1

 
The percentage of subjects with less than a high school education was higher in women than in men and in subjects with a higher BMI at age 25 -y (≥25) than in leaner participants. A higher prevalence of ex-smoking and current smoking was found in men than in women. There were more never and fewer ever smokers in subjects with a BMI < 23 at age 25 y compared with men and women with a BMI ≥ 23.0.

BMIs at ages 25, 40, and 55 y were strongly correlated: BMI at age 25 y –BMI at age 40 y (r = 0.72), BMI at age 40 y –BMI at age 55 y (r = 0.60), and BMI at age 25 y –BMI at age 55 y (r = 0.40). BMI changes between ages 25 and 40 y, 40 and 55 y, and 25 and 55 y were slightly inversely associated with BMI at age 25 y (r = –0.14, –0.04, and –0.10, respectively).

Interaction terms for weight change and sex were significant (P < 0.05); thus, we stratified all subsequent data analyses for men and women. The associations of BMI at age 25 y, BMI changes between ages 25 and 40 y, and BMI changes between ages 40 and 55 y with risk of type 2 diabetes are shown in Table 2Go. After multivariate adjustment, the point estimates of relative risk (RR) of type 2 diabetes was slightly higher for BMI gains between ages 25 and 40 y than between ages 40 and 55 y (P < 0.0001). A 1-unit higher BMI at the age of 25 y was associated with an RR of 1.15 (95% CI: 1.11, 1.19) for men and of 1.11 (95% CI: 1.07, 1.15) for women, independent of subsequent weight changes. BMI at baseline was associated with an RR of 1.21 (95% CI: 1.18, 1.24) for men and of 1.15 (95% CI: 1.13, 1.17) for women for a 1-unit increase (data not shown). We repeated our analysis after excluding all subjects who were younger than 50 y at the time of the baseline examination. The results remained similar; however, BMI change between ages 40 and 55 y was not significantly differently related to diabetes risk compared with BMI change between age 25 and 40 y among men (P < 0.19), whereas this significance persisted among women (P < 0.02). Slightly lower point estimates of diabetes risk for body weight gain in early and later adulthood were observed for men aged 25–40 y (RR: 1.24; 95% CI: 1.18, 1.30) and 40–55 y (RR: 1.19; 95% CI: 1.15, 1.24) and for women aged 25–40 y (RR: 1.22; 95% CI: 1.17, 1.27) and 40–55 y (RR: 1.14; 95% CI: 1.10, 1.19).


View this table:
[in this window]
[in a new window]
  		TABLE 2 Relative risk (RR) and 95% CI of type 2 diabetes for BMI change (in kg/m2) by components of BMI history1

 
The following example illustrates clearly that an initial weight change between ages 25 and 40 y is associated with a higher risk of type 2 diabetes than is a subsequent weight change between ages 40 and 55 y: A physician has 2 male patients aged 55 y with a BMI of 30. At the age of 25 y, both patients had a BMI of 25, but they gained weight (5 kg/m2) in different decades. One had gained weight between ages 25 and 40 y of age, and the other had a recent increase in BMI between ages 40 and 55 y. Compared with a man with a similar BMI at age 25 y, but with a stable weight until age 55 y, the RR of the man who gained weight initially was 3.1 (calculated as 1.255), ie, 1.7 times the RR of subsequent weight gain between ages 40 and 55 y (1.135 = 1.8).

We further analyzed the risk of different weight change histories by categorizing the BMI changes into 3 groups (loss or stable, moderate gain, and severe gain) (Table 3Go). Compared with men and women with weight loss or stable weight during both 15-y time periods of early and later adulthood, moderate or severe BMI gain during either or both periods was associated with an increased risk of diabetes. In detail, severe weight gain between 25 and 40 y of age and stable weight management between 40 and 55 y of age was associated with a diabetes risk in men and women that was 1.5 times and 4.3 times, respectively, the risk in those with stable weight management in early adulthood and severe gains in later life.


View this table:
[in this window]
[in a new window]
  		TABLE 3 Relative risk (RR) and 95% CI of type 2 diabetes by BMI change (in kg/m2) in men and women1

 
We examined the weight change groups loss or stable, moderate gain, and severe gain stratified for normal weight and overweight subjects at age 25 y (data not shown). The reference group was persons with a BMI < 25 at age 25 y and subsequent successful weight management during adulthood. There was no increased risk for men and women who were overweight or obese at age 25 y who had lost weight or had stable weight during each time period compared with normal-weight subjects whose weight was stable until age 55 y. However, men and women who gained weight during one or both time periods generally tended to have a higher risk if they were overweight or obese at age 25 y compared with normal-weight subjects with similar weight changes.

In addition, we observed an inverse association between weight change during ages 25–40 y and age at diagnosis, independent of subsequent weight changes. The mean (±SD) age at diabetes diagnosis was lowest in men and women with severe weight gain (men: 56.4 y ± 7.0; women: 59.1 y ± 7.4) and highest in subjects with weight loss or stable weight (men: 61.7 y ± 6.0; women: 62.2 y ± 6.6); the difference between groups was 5 y in men and 3 y in women. Men and women with a normal BMI at age 25 y and stable weight management during early adulthood had the latest age at diabetes diagnosis (men: 63.7 y ± 4.4; women: 61.4 y ± 6.7), whereas overweight subjects with a severe BMI gain between ages 25 and 40 y had, on average, a 6-y earlier onset of diabetes (men: 55.5 y ± 7.0; women: 56.2 y ± 7.0). Generally, earlier weight gain during adulthood was related to a higher diabetes risk than was later weight gain in both normal and overweight participants. Overweight in men and women at age 25 y resulted in a 2-y earlier age at diabetes diagnosis compared with normal-weight subjects. Age at diabetes diagnosis did not differ significantly between younger (<52.7 y) and older (≥52.7 y) men and women.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
We found that weight gain—measured as the change in BMI—in men and women in early adulthood (ages 25–40 y) was more strongly associated with risk of type 2 diabetes than was a subsequent increase in BMI. Also, greater weight gain, particularly in early adulthood, resulted in a younger age at onset of diabetes.

As in our study, weight gain between age 18 y and the time of the baseline assessment in 1976 was associated with a 1.9 (95% CI: 1.5, 2.3) higher risk for a weight gain of 5.0–7.9 kg and a 2.7 (95% CI: 2.1, 3.3) higher risk for a weight gain of 8.0–10.9 kg, independent of BMI at age 18 y for women in the Nurses' Health Study (10). The Health Professionals Follow-Up study indicated that a weight gain since age 21 y of 8–9 kg increased the risk of diabetes by 3.5 (95% CI: 2.0, 6.3) (7) and for every 1-kg of weight gained the risk increased by 7.3% (22). BMI at age 18 y for women or at age 21 y for men and weight gain since that age as well as an adult weight gain of >4.5 kg (40–60 y of age) and weight gain in general (between 25 and 50 y of age) were already identified as predictors of the risk of acquiring type 2 diabetes in later adulthood (2, 7, 22, 37–39). As in our study, body weight was assessed retrospectively in these previous studies in early adulthood. Similar results were obtained when BMI was self-reported in a prospective longitudinal study among former male medical students aged 20–29 y at baseline who were followed over 15 y, and body weight changes during this period were evaluated with regard to the risk of developing type 2 diabetes in middle age (12). Overweight at age 25 y was a strong predictor of diabetes in middle age, and BMI at 25, 35, and 45 y of age was strongly associated with diabetes risk.

However, to our knowledge, the effect of different weight-history components, comparing early (25–40 y) and later adulthood (40–55 y), risk of type 2 diabetes, as well as the effect of weight change on age at diabetes onset has not been investigated in previous studies. Our major finding with respect to early weight gain is noteworthy regarding 2 aspects.

First, we observed that weight gain in early adulthood (25–40 y) strongly predicts the risk of type 2 diabetes with a latency period of {approx}15–23 y. The stronger association with weight gain in earlier adulthood than in later adulthood might be explained by the longer duration of exposure to cumulative excessive body fat. The time at risk of type 2 diabetes under the exposure of increased body weight is higher for men and women with persistent obesity. That the duration of obesity is a significant risk factor for type 2 diabetes, independently of current degree of obesity, was observed in previous studies (6, 19, 23). One explanation for the higher risks among women with moderate or severe BMI gain during early adulthood than in men could be the relation between parity and obesity. The childbearing years have been identified as a critical period for substantial excess weight gain and development of obesity, which cannot be explained by behavioral changes (40–42). A retrospective analysis in women indicated a significant association between parity and age at diabetes diagnosis, which suggests that pregnancy might be a promoter of developing diabetes via the obesity pathway (43).

Second, weight gain between 25 and 40 y of age resulted in an earlier onset of type 2 diabetes. The difference in age at diabetes diagnosis for stable weight compared with severe weight gain was 5 y for men and 3 y for women. To develop type 2 diabetes, more obese subjects might require a shorter duration of obesity. We found only one epidemiologic study that investigated the association between obesity and age of onset of diabetes, which indicated an inverse relation between both (24). The average BMI was significantly higher for subjects with type 2 diabetes diagnosed before age 45 y than for diabetes diagnosed in older age (P < 0.001). The average BMI decreased from 38.3 in the youngest age group to 28.8 in the oldest age group (P for trend <0.0001 for trend). In our study, no difference in age at diabetes diagnosis could be found between younger and older subjects. A severe gain in early adulthood resulted in an earlier age of diagnosis after a longer, 2 decades-long latency period.

Nevertheless, some limitations of this study should be acknowledged. We considered only clinically apparent type 2 diabetes. Our results should not have been biased by disease misclassification, because all potential cases were verified through medical records. Given the resulting high predictive value positive of the disease classification, the remaining misclassification (nonidentified cases) should not have biased the estimated risk (44).

One limitation was that information on body weight at ages 25 and 40 y was self-reported, whereas weight and height at baseline were measured by trained staff. Although self-reported retrospective body weight has been shown to be highly correlated with measured weight by technicians, underreporting of past weight in overweight participants and overreporting in underweight participants may have occurred (45–50). Furthermore, body weight at age 40 y was only tested for reproducibility and not for validity (32). However, previous studies investigated the validity of self-reported past body weight, which indicated a high level of accuracy compared with measured body weight (48, 51). Together, this could have resulted in an over- or underestimation of BMI gain in past time periods. Another uncertainty is that we did not know the detailed characteristics of body change over 15 y in early life, eg, whether a subject gained weight continuously or whether weight cycling occurred over the time period. We assumed that total weight change indicated a continuous change in body weight, irrespective of type of change. A second limitation was that we calculated BMI at ages 25, 40, and 55 y and measured height at baseline only. Previous studies showed a decrease in body height in advanced age (52), which would result in a greater BMI given a stable weight. However, because only 3.2% of the cohort were aged ≥65 y and the mean age of the cohort was 53 y, a change in height in that period of life seems to be unlikely and a possible bias can be neglected (53). Third, retrospective data on body fat distribution, particularly waist and hip circumferences, were not available in our cohort; thus, the strong association between obesity and type 2 diabetes might have been underestimated when BMI was used as the only anthropometric variable (54). Furthermore, sporting activity in the 12 mo before the baseline examination was considered in our analysis, because information on physical activity throughout adult life was not available. Thus, we might not have been able to sufficiently control for confounding by activity.

To our knowledge, this is the first study to indicate that weight gain in early life results in an earlier age at diagnosis of type 2 diabetes after a latency period of 15–23 y. However, it should be considered that the age at diabetes diagnosis does not necessarily reflect the time of onset of diabetes. Subjects who develop type 2 diabetes in adult age show symptoms of hyperglycemia, insulin resistance, and impaired glucose tolerance years before manifestation of the onset of type 2 diabetes (55). Because we verified all self-reported cases of type 2 diabetes from medical records, it is possible that subjects with regular contact with medical personnel have a higher probability of early detection of diabetes than do those with less frequent contact with medical personnel. Therefore, the importance of the age effect at diabetes diagnosis in our analysis may have been underestimated, and further research is needed.

In summary, even modest weight gain in adult life is associated with a substantial risk of developing type 2 diabetes, and moderate or severe weight gain in early life is a stronger risk factor for diabetes than is weight gain after age 40 y. For relevance in clinical practice, it might be important to measure the current BMI and to assess comprehensively the history of weight gain or weight loss in earlier decades of adult life to provide a more precise risk profile for the patient. After a latency period of 15–23 y, age at diabetes diagnosis is earlier for overweight subjects and for those with a weight gain >3–12 kg in younger adulthood than in normal-weight subjects and those with marginal weight changes. This finding stresses the importance of maintaining a healthy body weight throughout adult life.


   ACKNOWLEDGMENTS
 
We thank Wolfgang Bernigau for statistical support and E Kohlsdorf for data management.

AS and HB designed the analysis plan. AS conducted the statistical analysis. AS, MBS, KH, AK, and HB interpreted the data. AS and MBS drafted the manuscript. KH provided significant consultation on statistical analysis. KH, AK, and HB revised the manuscript. None of the authors had any conflicts of interest.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Wilson PW, Anderson KM, Kannel WB. Epidemiology of diabetes mellitus in the elderly. The Framingham Study. Am J Med 1986; 80: 3–9.
  2. Holbrook TL, Barrett-Connor E, Wingard DL. The association of lifetime weight and weight control patterns with diabetes among men and women in an adult community. Int J Obes 1989; 13: 723–9.[Medline]
  3. Lundgren H, Bengtsson C, Blohme G, Lapidus L, Sjostrom L. Adiposity and adipose tissue distribution in relation to incidence of diabetes in women: results from a prospective population study in Gothenburg, Sweden. Int J Obes 1989; 13: 413–23.[Medline]
  4. Kaye SA, Folsom AR, Sprafka JM, Prineas RJ, Wallace RB. Increased incidence of diabetes mellitus in relation to abdominal adiposity in older women. J Clin Epidemiol 1991; 44: 329–34.[Medline]
  5. Haffner SM, Mitchell BD, Hazuda HP, Stern MP. Greater influence of central distribution of adipose tissue on incidence of non-insulin-dependent diabetes in women than men. Am J Clin Nutr 1991; 53: 1312–7.[Abstract/Free Full Text]
  6. Everhart JE, Pettitt DJ, Bennett PH, Knowler WC. Duration of obesity increases the incidence of NIDDM. Diabetes 1992; 41: 235–40.[Abstract]
  7. Chan JM, Rimm EB, Colditz GA, Stampfer MJ, Willett WC. Obesity, fat distribution, and weight gain as risk factors for clinical diabetes in men. Diabetes Care 1994; 17: 961–9.[Abstract]
  8. Perry IJ, Wannamethee SG, Walker MK, Thomson AG, Whincup PH, Shaper AG. Prospective study of risk factors for development of non-insulin dependent diabetes in middle aged British men. BMJ 1995; 310: 560–4.[Abstract/Free Full Text]
  9. Hanson RL, Narayan KM, McCance DR, et al. Rate of weight gain, weight fluctuation, and incidence of NIDDM. Diabetes 1995; 44: 261–6.[Abstract]
  10. Colditz GA, Willett WC, Rotnitzky A, Manson JE. Weight gain as a risk factor for clinical diabetes mellitus in women. Ann Intern Med 1995; 122: 481–6.[Abstract/Free Full Text]
  11. Wannamethee SG, Shaper AG, Walker M. Overweight and obesity and weight change in middle aged men: impact on cardiovascular disease and diabetes. J Epidemiol Community Health 2005; 59: 134–9.[Abstract/Free Full Text]
  12. Brancati FL, Wang NY, Mead LA, Liang KY, Klag MJ. Body weight patterns from 20 to 49 years of age and subsequent risk for diabetes mellitus: the Johns Hopkins Precursors Study. Arch Intern Med 1999; 159: 957–63.[Abstract/Free Full Text]
  13. Cassano PA, Rosner B, Vokonas PS, Weiss ST. Obesity and body fat distribution in relation to the incidence of non- insulin-dependent diabetes mellitus. A prospective cohort study of men in the normative aging study. Am J Epidemiol 1992; 136: 1474–86.
  14. Field AE, Coakley EH, Must A, et al. Impact of overweight on the risk of developing common chronic diseases during a 10-year period. Arch Intern Med 2001; 161: 1581–6.[Abstract/Free Full Text]
  15. Ford ES, Williamson DF, Liu S. Weight change and diabetes incidence: findings from a national cohort of US adults. Am J Epidemiol 1997; 146: 214–22.[Abstract/Free Full Text]
  16. Haffner SM, Stern MP, Mitchell BD, Hazuda HP, Patterson JK. Incidence of type II diabetes in Mexican Americans predicted by fasting insulin and glucose levels, obesity, and body-fat distribution. Diabetes 1990; 39: 283–8.[Abstract]
  17. Ishikawa-Takata K, Ohta T, Moritaki K, Gotou T, Inoue S. Obesity, weight change and risks for hypertension, diabetes and hypercholesterolemia in Japanese men. Eur J Clin Nutr 2002; 56: 601–7.[Medline]
  18. Resnick HE, Valsania P, Halter JB, Lin X. Relation of weight gain and weight loss on subsequent diabetes risk in overweight adults. J Epidemiol Community Health 2000; 54: 596–602.[Abstract/Free Full Text]
  19. Wannamethee SG, Shaper AG. Weight change and duration of overweight and obesity in the incidence of type 2 diabetes. Diabetes Care 1999; 22: 1266–72.[Abstract]
  20. Carey VJ, Walters EE, Colditz GA, et al. Body fat distribution and risk of non-insulin-dependent diabetes mellitus in women. The Nurses' Health Study. Am J Epidemiol 1997; 145: 614–9.
  21. Sakurai Y, Teruya K, Shimada N, et al. Relationship between weight change in young adulthood and the risk of NIDDM. The Sotetsu Study. Diabetes Care 1997; 20: 978–82.
  22. Koh-Banerjee P, Wang Y, Hu FB, Spiegelman D, Willett WC, Rimm EB. Changes in body weight and body fat distribution as risk factors for clinical diabetes in US men. Am J Epidemiol 2004; 159: 1150–9.[Abstract/Free Full Text]
  23. Sakurai Y, Teruya K, Shimada N, et al. Association between duration of obesity and risk of non-insulin- dependent diabetes mellitus. The Sotetsu Study. Am J Epidemiol 1999; 149: 256–60.
  24. Hillier TA, Pedula KL. Characteristics of an adult population with newly diagnosed type 2 diabetes: the relation of obesity and age of onset. Diabetes Care 2001; 24: 1522–7.[Abstract/Free Full Text]
  25. Abdul-Ghani MA, Kher J, Abbas N, Najami T. Association of high body mass index with low age of disease onset among Arab women with type 2 diabetes in a primary care clinic. Isr Med Assoc J 2005; 7: 360–3.[Medline]
  26. Bergmann KE, Mensink GB. [Anthropometric data and obesity]. Gesundheitswesen 1999; 61: S115–20(in German).
  27. Riboli E, Hunt KJ, Slimani N, et al. European Prospective Investigation into Cancer and Nutrition (EPIC): study populations and data collection. Public Health Nutr 2002; 5: 1113–24.[Medline]
  28. Boeing H, Korfmann A, Bergmann MM. Recruitment procedures of EPIC-Germany. European Investigation into Cancer and Nutrition. Ann Nutr Metab 1999; 43: 205–15.
  29. Bergmann MM, Bussas U, Boeing H. Follow-up procedures in EPIC-Germany–data quality aspects. European Prospective Investigation into Cancer and Nutrition. Ann Nutr Metab 1999; 43: 225–34.
  30. Kroke A, Bergmann MM, Lotze G, Jeckel A, Klipstein-Grobusch K, Boeing H. Measures of quality control in the German component of the EPIC study. European Prospective Investigation into Cancer and Nutrition. Ann Nutr Metab 1999; 43: 216–24.
  31. Klipstein-Grobusch K, Georg T, Boeing H. Interviewer variability in anthropometric measurements and estimates of body composition. Int J Epidemiol 1997; 26: S174–80.[Abstract/Free Full Text]
  32. Klipstein-Grobusch K, Kroke A, Boeing H. Reproducibility of self-reported past body weight. Eur J Clin Nutr 1998; 52: 525–8.[Medline]
  33. Klipstein-Grobusch K, Becker N, Kroke A, Boeing H. Patterns of past alcohol consumption in the EPIC-Germany cohorts. European Investigation into Cancer and Nutrition. Ann Nutr Metab 1999; 43: 258–65.
  34. De Stavola BL, Nitsch D, dos Santos Silva I, et al. Statistical issues in life course epidemiology. Am J Epidemiol 2006; 163: 84–96.[Abstract/Free Full Text]
  35. WHO. Obesity: preventing and managing the global epidemic. Report of a WHO consultation on obesity. Geneva, Switzerland: World Health Organization, 2000: 265.
  36. USDA/USDHHS. Report of the Dietary Guidelines Advisory Committee on the Dietary Guidelines for Americans. 5th ed. Washington, DC: US Department of Agriculture and US Department of Health and Human Services, 2000.
  37. Colditz GA, Willett WC, Stampfer MJ, et al. Weight as a risk factor for clinical diabetes in women. Am J Epidemiol 1990; 132: 501–13.[Abstract/Free Full Text]
  38. Harris MI. Impaired glucose tolerance in the U.S. population. Diabetes Care 1989; 12: 464–74.[Abstract]
  39. Oguma Y, Sesso HD, Paffenbarger RS Jr, Lee IM. Weight change and risk of developing type 2 diabetes. Obes Res 2005; 13: 945–51.[Medline]
  40. Smith DE, Lewis CE, Caveny JL, Perkins LL, Burke GL, Bild DE. Longitudinal changes in adiposity associated with pregnancy. The CARDIA Study. Coronary Artery Risk Development in Young Adults Study. JAMA 1994; 271: 1747–51.
  41. Williamson DF, Madans J, Pamuk E, Flegal KM, Kendrick JS, Serdula MK. A prospective study of childbearing and 10-year weight gain in US white women 25 to 45 years of age. Int J Obes Relat Metab Disord 1994; 18: 561–9.[Medline]
  42. Gunderson EP, Murtaugh MA, Lewis CE, Quesenberry CP, West DS, Sidney S. Excess gains in weight and waist circumference associated with childbearing: The Coronary Artery Risk Development in Young Adults Study (CARDIA). Int J Obes Relat Metab Disord 2004; 28: 525–35.[Medline]
  43. Cheung NW. Is parity associated with earlier diagnosis of type 2 diabetes? Diabetes Res Clin Pract 2004; 66: 287–91.[Medline]
  44. Greenland S. Basic methods for sensitivity analysis and external adjustment. In: Rothman K, Greenland S, eds. Modern epidemiology. 2nd ed. Philadelphia, PA: Lippincott-Raven, 1998.
  45. Tamakoshi K, Yatsuya H, Kondo T, et al. The accuracy of long-term recall of past body weight in Japanese adult men. Int J Obes Relat Metab Disord 2003; 27: 247–52.[Medline]
  46. Spencer EA, Appleby PN, Davey GK, Key TJ. Validity of self-reported height and weight in 4808 EPIC-Oxford participants. Public Health Nutr 2002; 5: 561–5.[Medline]
  47. Kuskowska-Wolk A, Bergstrom R, Bostrom G. Relationship between questionnaire data and medical records of height, weight and body mass index. Int J Obes Relat Metab Disord 1992; 16: 1–9.[Medline]
  48. Stevens J, Keil JE, Waid LR, Gazes PC. Accuracy of current, 4-year, and 28-year self-reported body weight in an elderly population. Am J Epidemiol 1990; 132: 1156–63.[Abstract/Free Full Text]
  49. Rowland ML. Self-reported weight and height. Am J Clin Nutr 1990; 52: 1125–33.[Abstract/Free Full Text]
  50. Perry GS, Byers TE, Mokdad AH, Serdula MK, Williamson DF. The validity of self-reports of past body weights by U.S. adults. Epidemiology 1995; 6: 61–6.[Medline]
  51. Stunkard AJ, Albaum JM. The accuracy of self-reported weights. Am J Clin Nutr 1981; 34: 1593–9.[Abstract/Free Full Text]
  52. Suominen H. Changes in physical characteristics and body composition during 5-year follow-up in 75- and 80-year-old men and women. Scand J Soc Med Suppl 1997; 53: 19–24.[Medline]
  53. Bartali B, Benvenuti E, Corsi AM, et al. Changes in anthropometric measures in men and women across the life- span: findings from the InCHIANTI study. Soz Praventivmed 2002; 47: 336–48.[Medline]
  54. Wang Y, Rimm EB, Stampfer MJ, Willett WC, Hu FB. Comparison of abdominal adiposity and overall obesity in predicting risk of type 2 diabetes among men. Am J Clin Nutr 2005; 81: 555–63.[Abstract/Free Full Text]
  55. Martin BC, Warram JH, Krolewski AS, Bergman RN, Soeldner JS, Kahn CR. Role of glucose and insulin resistance in development of type 2 diabetes mellitus: results of a 25-year follow-up study. Lancet 1992; 340: 925–9.[Medline]
Received for publication December 20, 2005. Accepted for publication March 24, 2006.





This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Schienkiewitz, A.
Right arrow Articles by Boeing, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Schienkiewitz, A.
Right arrow Articles by Boeing, H.
Agricola
Right arrow Articles by Schienkiewitz, A.
Right arrow Articles by Boeing, H.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS